End-to-end aspect category sentiment analysis based on type graph convolutional networks

For the existing aspect category sentiment analysis research,most of the aspects are given for sentiment extraction,and this pipeline method is prone to error accumulation,and the use of graph convolutional neural network for aspect category sentiment analysis does not fully utilize the dependency type information between words,so it cannot enhance feature extraction.This paper proposes an end-to-end aspect category sentiment analysis(ETESA)model based on type graph convolutional networks.The model uses the bidirectional encoder representation from transformers(BERT)pretraining model to obtain aspect categories and word vectors containing contextual dynamic semantic information,which can solve the problem of polysemy;when using graph convolutional network(GCN)for feature extraction,the fusion operation of word vectors and initialization tensor of dependency types can obtain the importance values of different dependency types and enhance the text feature representation;by transforming aspect category and sentiment pair extraction into multiple single-label classification problems,aspect category and sentiment can be extracted simultaneously in an end-to-end way and solve the problem of error accumulation.Experiments are tested on three public datasets,and the results show that the ETESA model can achieve higher Precision,Recall and F1 value,proving the effectiveness of the model.

Four Types of Percolation Transitions in the Cluster Aggregation Network Model

We study the percolation transition in a one-species cluster aggregation network model, in which the parameter α describes the suppression on the cluster sizes. It is found that the model can exhibit four types of percolation transitions, two continuous percolation transitions and two discontinuous ones. Continuous and discontinuous percolation transitions can be distinguished from each other by the largest single jump. Two types of continuous percolation transitions show different behaviors in the time gap. Two types of discontinuous percolation transitions are different in the time evolution of the cluster size distribution. Moreover, we also find that the time gap may also be a measure to distinguish different discontinuous percolations in this model.

The essential order of approximation for nearly exponential type neural networks

For the nearly exponential type of feedforward neural networks （neFNNs）, it is revealed the essential order of their approximation. It is proven that for any continuous function defined on a compact set of Rd, there exists a three-layer neFNNs with fixed number of hidden neurons that attain the essential order. When the function to be approximated belongs to the α-Lipschitz family （0 〈α≤ 2）, the essential order of approxi- mation is shown to be O（n^-α） where n is any integer not less than the reciprocal of the predetermined approximation error. The upper bound and lower bound estimations on approximation precision of the neFNNs are provided. The obtained results not only characterize the intrinsic property of approximation of the neFNNs, but also uncover the implicit relationship between the precision （speed） and the number of hidden neurons of the neFNNs.

Relationship between atomic structure and crystal type studied by artificial neural network

We have used chemical bond parameters and pattern recognition method to investigatethe regularities of the crystal type of alloy phase,and achieved good results.Theparameters used,however,are semi-empirical paramters,which are not very strict fromtheoretical viewpoint.In this letter,we use the numbers describing atomic structure(thenumbers of valence electrons Z1,Z2,the principal quantum numbers of valence electrons n1,

Full Metal Species Quantification of Metal Supported Catalysts Through Massive TEM Images Recognition

For a practical high-loading single-atom catalyst,it is prone to forming diverse metal species owing to either the synthesis inhomogeneity or the reaction induced aggregation.The diversity of this metal species challenges the discerning about the contributions of specific metal species to the catalytic performance,and thus hampers the rational catalyst design.In this paper,a distinct solution of dispersion analysis based on transmission electron microscopy imaging specialized for metal-supported catalysts has been proposed in the capability of full-metal-species quantification(FMSQ)from single atoms to nanoparticles,including dispersion densities,shape geometry,and crystallographic surface exposure.This solution integrates two image-recognition algorithms including the electron microscopy-based atom recognition statistics(EMARS)for single atoms and U-Net type deep learning network for nanoparticles in different shapes.When applied to the C_(3)N_(4)-and nitrogen-doped carbon-supported catalysts,the FMSQ method successfully identifies the specific activity contributions of Au single atoms and particles in butadiene hydrogenation,which presents remarkable variation with the metal species constitution.This work demonstrates a promising value of our FMSQ strategy for identifying the activity origin of heterogeneous catalysis.

The research of traffic density extraction method under vehicular ad hoc network environment

Purpose–Traffic density is one of the most important parameters to consider in the traffic operationfield.Owing to limited data sources,traditional methods cannot extract traffic density directly.In the vehicular ad hoc network(VANET)environment,the vehicle-to-vehicle(V2V)and vehicle-to-infrastructure(V2I)interaction technologies create better conditions for collecting the whole time-space and refined traffic data,which provides a new approach to solving this problem.Design/methodology/approach–On that basis,a real-time traffic density extraction method has been proposed,including lane density,segment density and network density.Meanwhile,using SUMO and OMNet11 as traffic simulator and network simulator,respectively,the Veins framework as middleware and the two-way coupling VANET simulation platform was constructed.Findings–Based on the simulation platform,a simulated intersection in Shanghai was developed to investigate the adaptability of the model.Originality/value–Most research studies use separate simulation methods,importing trace data obtained by using from the simulation software to the communication simulation software.In this paper,the tight coupling simulation method is applied.Using real-time data and history data,the research focuses on the establishment and validation of the traffic density extraction model.